It's Time to Give the Final Final Exam
My oldest son is in ninth grade. I've written before about his eventful journey through school so far, and about some of the things that have made it more difficult than it should have been. This year, since he's a freshman in high school, he's taking final exams for the first time. Ah, the venerable final exam.
Who doesn't love the final exam? Wait, let me rephrase that: who loves the final exam? I can remember giving final exams as a high school teacher, when I had to, but I stopped giving them a long time ago. As teachers, we have it drilled into our heads when we're learning to teach that summative assessment is one of the many important kinds of assessment we have to do in order to ensure that we know that students are actually learning all the wonderful things we taught them, but is it really? Together with formative assessment and diagnostic assessment, summative assessment forms the holy trinity of student assessment. Or maybe it's a three-legged stool. I can't remember which. What I do know is that holy trinities are sacred and we should have faith in them and not ask too many questions about them. Three legged stools, on the other hand, can be really hard to sit on if they aren't built properly, or if one of the legs is weaker than the other ones.
Too many final exams make me think of three-legged stools, and rickety ones at that. This is mainly because of that summative assessment leg. To me, final exams represent much of what's wrong with our approach to education these days. In our relentless drive to hold everyone accountable for what happens in schools the emphasis on summative assessment—on that final test of whether or not the material has been mastered or the outcomes have been achieved or the benchmarks have been met or whatever; the one that spits out that One Special Number or Letter that clears everything up—has acquired an importance that outweighs its value. It's one of those things we do because, well, it's one of those things we do.
There are better ways to measure student learning. For starters, every teacher who knows his or her business knows that one of the major shortcomings of big final tests is that they only offer a snapshot in time of actual student learning. The timing isn't ideal either: in our local high school, like in so many others, this is the last big thing to do before the new semester starts. That calls for a quick turnaround. And that means multiple choice. You know the problem with multiple choice, right? You can get the right answer without actually knowing anything. And you can get the wrong answer even if you know a lot, just for having the right answer to the wrong question.
This is to say nothing of the absurdity of giving the final after winter break, which is another thing we do here. I have an idea: let's have about eighteen weeks of uninterrupted instruction, then take two weeks off. Then let's take a major test that covers everything we've learned, and have it count for a major portion of your final grade. Better yet, let's take like five of them all in a row. Whose idea was this?
Most teachers I know don't give final exams because they want to—they do it because they have to. Somebody above them says so. Not only do we have to have accountability these days, but we also have to have it all perfectly aligned—or at least pretend we do. In my last years as a high school teacher we had gone from giving "county exams," which were apparently made using testmaking software somewhere in the central office complex, to having to give identical "benchmark exams" too. These were designed to keep everyone, in a district of over 100,000 kids, on the same schedule. Our kids always got crushed on those exams. It was a running joke, but not a funny one: I know I felt some pressure to help kids keep those class averages up in anticipation of a low grade on the trivia contest, and I'm sure I was not alone.
Final exams are usually built on the assumption that if you know everything you ought to be able to answer questions about anything. They almost never serve as a springboard to new learning—at best, they become a meaningless afterthought, a punctuation mark at the end of a school term that's quickly forgotten, and at worst they throw a whole semester's or year's worth of work into chaos. Worst of all, the need for final exams is totally obviated by the presence of other tests. Yet even as test schedules have become bloated beyond belief in the past two decades, the final exam continues to hang around—a relic that should remind us, if nothing else, that we were irrationally serious about assessing student learning even in the days before the made a federal case out of it.
Last week, my son got back his English exam score the same day he took it. I don't blame his teacher for this—I blame a school calendar that privileges final exams as a necessary rite of passage in high schools and a school culture that values simplistic assessment over more authentic and personalized evaluations of student learning. Summative assessments should actually be formative assessments—that is, they should lead from someplace to someplace else. It should take more than an hour or two to determine if 20 or 30 ninth graders have the reading, writing, and speaking skills they need to continue growing at the next level. It's natural for life to have starting and stopping points, but the overemphasis on finality prevents us from recognizing that continuity is important too.
You know why I stopped giving final exams? Because I could. At the post-secondary level, where teachers have real autonomy (even those of us trying to satisfy the capricious whims of bureaucrats in the state department of education), this is a choice—not a requirement. It's so clear now that students are taking too many tests that even President Obama—who spent much of his presidency helping to make the problem worse—admits there's too much testing going on. Here's one way to start turning the tide: let's kill the final exam.